Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus

Lukas Ortmann; Dawei Shi; Eyal Dassau; Francis J. Doyle; Steffen Leonhardt; Berno J.E. Misgeld

doi:10.1109/ASCC.2017.8287323

Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus

Lukas Ortmann
, Dawei Shi
, Eyal Dassau
, Francis J. Doyle
, Steffen Leonhardt
, Berno J.E. Misgeld

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

19 Citations (Scopus)

Abstract

The insulin sensitivity (IS) of the human body changes with a circadian rhythm. This adds to the time-varying feature of the glucose metabolism process and places challenges on the blood glucose (BG) control of patients with Type 1 Diabetes Mellitus. This paper presents a Model Predictive Controller that takes the periodic IS into account, in order to enhance BG control. The future effect of the IS is predicted using a machine learning technique, namely, a customized Gaussian Process (GP), based on historical training data. The training data for the GP is continuously updated during closed-loop control, which enables the control scheme to learn and adapt to intra-individual and inter-individual changes of the circadian IS rhythm. The necessary state information is provided by an Unscented Kalman Filter. The closed-loop performance of the proposed control scheme is evaluated for different scenarios (including fasting, announced meals and skipped meals) through in silico studies on simulation models of Göttingen Minipigs.

Original language	English
Title of host publication	2017 Asian Control Conference, ASCC 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1092-1097
Number of pages	6
ISBN (Electronic)	9781509015733
DOIs	https://doi.org/10.1109/ASCC.2017.8287323
Publication status	Published - 7 Feb 2018
Externally published	Yes
Event	2017 11th Asian Control Conference, ASCC 2017 - Gold Coast, Australia Duration: 17 Dec 2017 → 20 Dec 2017

Publication series

Name	2017 Asian Control Conference, ASCC 2017
Volume	2018-January

Conference

Conference	2017 11th Asian Control Conference, ASCC 2017
Country/Territory	Australia
City	Gold Coast
Period	17/12/17 → 20/12/17

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Artificial Pancreas
Gaussian Process
Model Predictive Control
insulin sensitivity

Access to Document

10.1109/ASCC.2017.8287323

Cite this

Ortmann, L., Shi, D., Dassau, E., Doyle, F. J., Leonhardt, S., & Misgeld, B. J. E. (2018). Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus. In 2017 Asian Control Conference, ASCC 2017 (pp. 1092-1097). (2017 Asian Control Conference, ASCC 2017; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ASCC.2017.8287323

@inproceedings{f621c73f7aa342b883e503d8d8f28e32,

title = "Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus",

abstract = "The insulin sensitivity (IS) of the human body changes with a circadian rhythm. This adds to the time-varying feature of the glucose metabolism process and places challenges on the blood glucose (BG) control of patients with Type 1 Diabetes Mellitus. This paper presents a Model Predictive Controller that takes the periodic IS into account, in order to enhance BG control. The future effect of the IS is predicted using a machine learning technique, namely, a customized Gaussian Process (GP), based on historical training data. The training data for the GP is continuously updated during closed-loop control, which enables the control scheme to learn and adapt to intra-individual and inter-individual changes of the circadian IS rhythm. The necessary state information is provided by an Unscented Kalman Filter. The closed-loop performance of the proposed control scheme is evaluated for different scenarios (including fasting, announced meals and skipped meals) through in silico studies on simulation models of G{\"o}ttingen Minipigs.",

keywords = "Artificial Pancreas, Gaussian Process, Model Predictive Control, insulin sensitivity",

author = "Lukas Ortmann and Dawei Shi and Eyal Dassau and Doyle, \{Francis J.\} and Steffen Leonhardt and Misgeld, \{Berno J.E.\}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 2017 11th Asian Control Conference, ASCC 2017 ; Conference date: 17-12-2017 Through 20-12-2017",

year = "2018",

month = feb,

day = "7",

doi = "10.1109/ASCC.2017.8287323",

language = "English",

series = "2017 Asian Control Conference, ASCC 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1092--1097",

booktitle = "2017 Asian Control Conference, ASCC 2017",

address = "United States",

}

Ortmann, L, Shi, D, Dassau, E, Doyle, FJ, Leonhardt, S & Misgeld, BJE 2018, Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus. in 2017 Asian Control Conference, ASCC 2017. 2017 Asian Control Conference, ASCC 2017, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 1092-1097, 2017 11th Asian Control Conference, ASCC 2017, Gold Coast, Australia, 17/12/17. https://doi.org/10.1109/ASCC.2017.8287323

Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus. / Ortmann, Lukas; Shi, Dawei; Dassau, Eyal et al.
2017 Asian Control Conference, ASCC 2017. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1092-1097 (2017 Asian Control Conference, ASCC 2017; Vol. 2018-January).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus

AU - Ortmann, Lukas

AU - Shi, Dawei

AU - Dassau, Eyal

AU - Doyle, Francis J.

AU - Leonhardt, Steffen

AU - Misgeld, Berno J.E.

PY - 2018/2/7

Y1 - 2018/2/7

N2 - The insulin sensitivity (IS) of the human body changes with a circadian rhythm. This adds to the time-varying feature of the glucose metabolism process and places challenges on the blood glucose (BG) control of patients with Type 1 Diabetes Mellitus. This paper presents a Model Predictive Controller that takes the periodic IS into account, in order to enhance BG control. The future effect of the IS is predicted using a machine learning technique, namely, a customized Gaussian Process (GP), based on historical training data. The training data for the GP is continuously updated during closed-loop control, which enables the control scheme to learn and adapt to intra-individual and inter-individual changes of the circadian IS rhythm. The necessary state information is provided by an Unscented Kalman Filter. The closed-loop performance of the proposed control scheme is evaluated for different scenarios (including fasting, announced meals and skipped meals) through in silico studies on simulation models of Göttingen Minipigs.

AB - The insulin sensitivity (IS) of the human body changes with a circadian rhythm. This adds to the time-varying feature of the glucose metabolism process and places challenges on the blood glucose (BG) control of patients with Type 1 Diabetes Mellitus. This paper presents a Model Predictive Controller that takes the periodic IS into account, in order to enhance BG control. The future effect of the IS is predicted using a machine learning technique, namely, a customized Gaussian Process (GP), based on historical training data. The training data for the GP is continuously updated during closed-loop control, which enables the control scheme to learn and adapt to intra-individual and inter-individual changes of the circadian IS rhythm. The necessary state information is provided by an Unscented Kalman Filter. The closed-loop performance of the proposed control scheme is evaluated for different scenarios (including fasting, announced meals and skipped meals) through in silico studies on simulation models of Göttingen Minipigs.

KW - Artificial Pancreas

KW - Gaussian Process

KW - Model Predictive Control

KW - insulin sensitivity

UR - https://www.scopus.com/pages/publications/85047488887

U2 - 10.1109/ASCC.2017.8287323

DO - 10.1109/ASCC.2017.8287323

M3 - Conference contribution

AN - SCOPUS:85047488887

T3 - 2017 Asian Control Conference, ASCC 2017

SP - 1092

EP - 1097

BT - 2017 Asian Control Conference, ASCC 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 11th Asian Control Conference, ASCC 2017

Y2 - 17 December 2017 through 20 December 2017

ER -

Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this